Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Malicious Package Detection research benchmarks · SOTA2 Research

Malicious Package Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

PyPI Malicious Package Dataset All Packages Combined (test)

97.08Accuracy

24

Jul 7, 2026

Cerebro RoBERTa

96.1Precision

17

Jun 23, 2026

QUT-DV25 (test)

99AUC

14

Apr 30, 2026

PyPI (Extremely Large ≥700KB)

96.33Accuracy

12

Jul 7, 2026

PyPI (Extremely Small ≤1.57KB)

95.96Accuracy

12

Jul 7, 2026

PyPI Extensive Imports (test)

95Accuracy

12

Jul 7, 2026

PyPI Minimal Imports (test)

96.6Accuracy

12

Jul 7, 2026

PyPI Malicious Package Dataset Large Packages N = 1162 (test)

96.89Accuracy

12

Jul 7, 2026

PyPI Malicious Package Dataset Small Packages N = 972 (test)

97.31Accuracy

12

Jul 7, 2026

Obfuscation Dataset

98.28Accuracy

11

Feb 26, 2026

97.37Accuracy

11

Feb 26, 2026

Evaluation Dataset

99.5Accuracy

11

Feb 26, 2026

Malicious package evaluation dataset

99.595Precision

10

Jun 18, 2026

99.5Accuracy

8

Apr 30, 2026

99.1Precision

7

Jun 2, 2026

NPM PyPI-trained (test)

48.8Precision

6

Jun 23, 2026

PyPI NPM-trained (test)

Cerebro RoBERTa

75.1Precision

6

Jun 23, 2026

MalwareBench adv (test)

86.81Recall (TPR) @ 1% FPR

5

Feb 27, 2026

MalwareBench (test)

95.64Recall @ 1% FPR

5

Feb 27, 2026

99.61Accuracy

3

Jun 23, 2026

Custom evaluation dataset

—

—Primary metric

0

Feb 26, 2026

Popular datasets

PyPI PyPI Malicious Package Dataset NPM MalwareBench D1

Follow for updates

Join SOTA2 Contact